Human trajectory prediction is essential and promising in many related applications. This is challenging due to the uncertainty of human behaviors, which can be influenced not only by himself, but also by the surround...Human trajectory prediction is essential and promising in many related applications. This is challenging due to the uncertainty of human behaviors, which can be influenced not only by himself, but also by the surrounding environment. Recent works based on long-short term memory(LSTM) models have brought tremendous improvements on the task of trajectory prediction. However, most of them focus on the spatial influence of humans but ignore the temporal influence. In this paper, we propose a novel spatial-temporal attention(ST-Attention) model,which studies spatial and temporal affinities jointly. Specifically,we introduce an attention mechanism to extract temporal affinity,learning the importance for historical trajectory information at different time instants. To explore spatial affinity, a deep neural network is employed to measure different importance of the neighbors. Experimental results show that our method achieves competitive performance compared with state-of-the-art methods on publicly available datasets.展开更多
Influenced by the layout of seismic network and the location of earthquakes,earthquake catalogs are often incomplete;such incompleteness of earthquake catalogue directly affects the analysis of sequence activity chara...Influenced by the layout of seismic network and the location of earthquakes,earthquake catalogs are often incomplete;such incompleteness of earthquake catalogue directly affects the analysis of sequence activity characteristics.In this paper,the GPU-acceleration-based g template matching method is used to scan the continuous waveforms of Chang Island earthquake swarm in Shandong Province from February 9 to August 20,2017.In total,15,286 earthquakes events were detected,which was more than 6 times compared with those in network catalogue and thus reduced the magnitude of completeness from 1.0 to 0.5.Based on the intergrated catalogue of earthquakes,the characteristics of Chang Island earthquake swarm were then analyzed using the Epidemic Type Aftershock Sequences(ETAS)model.The stochastic components in the ETAS model are used as a proxy for possible earthquake triggered by external forces(fluids).The results show that the proportion of earthquakes triggered by external forces of Chang Island swarm increases gradually(from 31.9%to 63.5%)and then decreases.The latter stage of swarm development is mainly affected by the self-excitation of earthquakes,suggesting that the fluids play an important role in the development of the Chang Island swarm.However,the triggering intensity of fluids to microseismicity is divergent in different periods,which may be related to the process of fluid permeation.展开更多
Accurate traffic flow prediction has a profound impact on modern traffic management. Traffic flow has complex spatial-temporal correlations and periodicity, which poses difficulties for precise prediction. To address ...Accurate traffic flow prediction has a profound impact on modern traffic management. Traffic flow has complex spatial-temporal correlations and periodicity, which poses difficulties for precise prediction. To address this problem, a Multi-head Self-attention and Spatial-Temporal Graph Convolutional Network (MSSTGCN) for multiscale traffic flow prediction is proposed. Firstly, to capture the hidden traffic periodicity of traffic flow, traffic flow is divided into three kinds of periods, including hourly, daily, and weekly data. Secondly, a graph attention residual layer is constructed to learn the global spatial features across regions. Local spatial-temporal dependence is captured by using a T-GCN module. Thirdly, a transformer layer is introduced to learn the long-term dependence in time. A position embedding mechanism is introduced to label position information for all traffic sequences. Thus, this multi-head self-attention mechanism can recognize the sequence order and allocate weights for different time nodes. Experimental results on four real-world datasets show that the MSSTGCN performs better than the baseline methods and can be successfully adapted to traffic prediction tasks.展开更多
In order to reveal the relationship between water injection in mine wells in the Zigong area and seismicity, we divide the historical earthquakes of ML ≥ 1.2 into 3 phases according to seismicity behavior, and the ET...In order to reveal the relationship between water injection in mine wells in the Zigong area and seismicity, we divide the historical earthquakes of ML ≥ 1.2 into 3 phases according to seismicity behavior, and the ETAS model parameters are then inversed by the POWELL method. The results show that phase 1 and 2, in which there is no water injection, have moderate-to-low ratio of background earthquakes （40 % - 50 % ）, and aftershocks are relatively less for a single earthquake sequence. In phase 3, where there is water injection, the aftershocks triggered by foreshocks dominate （ 93. 1% ）, and background earthquakes amount only to 6. 9 %, less than those of phase 1 and 2. The results conflict with the existing cognition. To resolve this problem, we propose that the occurrence ratio of background earthquakes in unit time, that is, the p value in ETAS model is used as an indicator of water injection triggered earthquakes. Compared to the first two phases, phase 3 has the largest u value, which illustrates that the water injection has an obvious triggering effect on earthquakes of this region.展开更多
利用基于时-空传染型余震序列(Epidemic Type Aftershock Sequence,简称ETAS)模型的随机除丛法,重新审视了2008年5月12日汶川M_S8.0地震前可能存在的长期地震活动异常,研究了川滇地区背景地震活动特征,并评估了当前的强震危险状态.对川...利用基于时-空传染型余震序列(Epidemic Type Aftershock Sequence,简称ETAS)模型的随机除丛法,重新审视了2008年5月12日汶川M_S8.0地震前可能存在的长期地震活动异常,研究了川滇地区背景地震活动特征,并评估了当前的强震危险状态.对川滇地区1 970年以来的M_L3.0以上的背景地震和丛集地震活动的研究结果表明,该地区地震丛集特征明显、时空分布很不均匀、地震序列常有前震事件.直接将概率值作为地震计数的权重,对地震丛集率空间分布图像分析表明,汶川M_S8.0地震前,龙门山断裂带中南段存在着长期、大范围的地震丛集率低值区,震前该段处于应力闭锁状态.对川滇地区地震从集率低值区内背景地震与全部地震的累积次数、b值和新定义的Δb等统计参量的分析表明,龙日坝与龙门山断裂带具有地震活动的关联性,川滇地区当前的强震潜在危险区可能是巧家地区和汶川M_S8.0地震破裂尚未穿越的龙门山断裂带南段.此外,还发现b值倾向于反映局部应力场变化,而△b能较为敏感地给出更大范围应力场的相对变化.展开更多
Macroscopic grasp of agricultural carbon emissions status, spatial-temporal characteristics as well as driving factors are the basic premise in further research on China’s agricultural carbon emissions. Based on 23 k...Macroscopic grasp of agricultural carbon emissions status, spatial-temporal characteristics as well as driving factors are the basic premise in further research on China’s agricultural carbon emissions. Based on 23 kinds of major carbon emission sources including agricultural materials inputs, paddy ifeld, soil and livestock breeding, this paper ifrstly calculated agricultural carbon emissions from 1995 to 2010, as well as 31 provinces and cities in 2010 in China. We then made a decomposed analysis to the driving factors of carbon emissions with logarithmic mean Divisia index （LMDI） model. The results show：（1） The amount of agricultural carbon emissions is 291.1691 million t in 2010. Compared with 249.5239 million t in 1995, it increased by 16.69%, in which, agricultural materials inputs, paddy ifeld, soil, enteric fermentation, and manure management accounted for 33.59, 22.03, 7.46, 17.53 and 19.39%of total agricultural carbon emissions, respectively. Although the amount exist ups and downs, it shows an overall trend of cyclical rise; （2） There is an obvious difference among regions：the amount of agricultural carbon emissions from top ten zones account for 56.68%, while 9.84%from last 10 zones. The traditional agricultural provinces, especially the major crop production areas are the main source regions. Based on the differences of carbon emission rations, 31 provinces and cities are divided into ifve types, namely agricultural materials dominant type, paddy ifeld dominant type, enteric fermentation dominant type, composite factors dominant type and balanced type. The agricultural carbon emissions intensity in west of China is the highest, followed by the central region, and the east zone is the lowest; （3） Compared with 1995, efifciency, labor and structure factors cut down carbon emissions by 65.78, 27.51 and 3.19%, respectively;while economy factor increase carbon emissions by 113.16%.展开更多
Set-nets are common alongshore fishing gear used in Haizhou Bay, which rely on flow to catch fish. The catch per unit effort(CPUE) of set-net is affected by spatial-temporal and environmental factors but no research h...Set-nets are common alongshore fishing gear used in Haizhou Bay, which rely on flow to catch fish. The catch per unit effort(CPUE) of set-net is affected by spatial-temporal and environmental factors but no research has been conducted on this subject. In this study, we used generalized additive models(GAMs) to explore the influence of spatial-temporal and environmental factors on CPUEs of species aggregated, small yellow croaker(Larimichthys polyactis), and octopus(Octopus variabilis) based on logbooks investigations conducted at 4 stations in an alongshore area of Haizhou Bay from 2011 to 2012. The results showed that all CPUEs exhibited significant spatial-temporal differences at various scales. Aggregated CPUE was high when the sea surface temperature(SST) was 15-18℃ and 20-23℃, which was mainly determined by life history traits of the octopus and small yellow croaker(optimal SSTs 14-17℃ and 19-24℃, respectively). Chlorophyll-a concentration had significant influences on the aggregated, small yellow croaker and octopus CPUEs at optimal ranges of 3.8-6.2 mg m^(-3), 4.2-4.8 mg m^(-3) and 4.5-5.5 mg m^(-3), respectively. Flow through the net had positive relationships with CPUEs. The approximate logarithmic trends in regression curves had a critical point of 2.5 Mm^3 d^(-1), which was the dividing point that differentiated whether the major factor affecting CPUEs was the flow velocity or the fishery resource. Our results from this study will help guide fishery production and improve catch rate of set-net fishing in Haizhou Bay.展开更多
To comprehensively understand the law of urban-rural relationship and propose scientific measures of urban-rural coordinated development in Northeast China,this study uses the coupling coordination degree model and ge...To comprehensively understand the law of urban-rural relationship and propose scientific measures of urban-rural coordinated development in Northeast China,this study uses the coupling coordination degree model and geographically and temporally weighted regression(GTWR)model to analyze the spatial-temporal patterns and the corresponding driving mechanisms of its urban-rural coordination since 1990.The results are as follows.First,the urban-rural coupling coordination degree in Northeast China was very low and improved slowly,but its stages of evolution is a good interpretation of the strategic arrangements of China's urbanization.Second,the urban-rural coupling coordination degree in Northeast China had spatial differences and was characterized by central polarization,converging on urban agglomeration,which was high in the south and low in the north.Moreover,the gap between the north and south weakened.Third,the spatial-temporal evolution of the urban-rural coordination relationship in Northeast China was influenced by pulling from the central cities,pushing from rural transformation,and government regulations.The influence intensity of the three mechanisms was weak,but the pulling from the central cities was stronger than that of the other two mechanisms.Furthermore,the spatial difference between the three mechanisms determines the spatial pattern and its evolution of the urban-rural coordination relationship in Northeast China.Fourth,to promote the development of urban-rural coordination in Northeast China,it is essential to advance urban-rural economic correlation,enhance the government^role in regulating and guiding,and adopt different policies for each region in Northeast China.展开更多
基金supported by the National Key Research and Development Program of China(2018AAA0101005,2018AAA0102404)the Program of the Huawei Technologies Co.Ltd.(FA2018111061SOW12)+1 种基金the National Natural Science Foundation of China(61773054)the Youth Research Fund of the State Key Laboratory of Complex Systems Management and Control(20190213)。
文摘Human trajectory prediction is essential and promising in many related applications. This is challenging due to the uncertainty of human behaviors, which can be influenced not only by himself, but also by the surrounding environment. Recent works based on long-short term memory(LSTM) models have brought tremendous improvements on the task of trajectory prediction. However, most of them focus on the spatial influence of humans but ignore the temporal influence. In this paper, we propose a novel spatial-temporal attention(ST-Attention) model,which studies spatial and temporal affinities jointly. Specifically,we introduce an attention mechanism to extract temporal affinity,learning the importance for historical trajectory information at different time instants. To explore spatial affinity, a deep neural network is employed to measure different importance of the neighbors. Experimental results show that our method achieves competitive performance compared with state-of-the-art methods on publicly available datasets.
基金sponsored by the National Key R&D Program of China(2016YFE0109300)the Seismological Science and Technology Spark Program(XH18026Y)+1 种基金Natural Science Foundation of Shandong Province(ZR2017QD014)Key R&D Program of Shandong Province(2016GSF120011)
文摘Influenced by the layout of seismic network and the location of earthquakes,earthquake catalogs are often incomplete;such incompleteness of earthquake catalogue directly affects the analysis of sequence activity characteristics.In this paper,the GPU-acceleration-based g template matching method is used to scan the continuous waveforms of Chang Island earthquake swarm in Shandong Province from February 9 to August 20,2017.In total,15,286 earthquakes events were detected,which was more than 6 times compared with those in network catalogue and thus reduced the magnitude of completeness from 1.0 to 0.5.Based on the intergrated catalogue of earthquakes,the characteristics of Chang Island earthquake swarm were then analyzed using the Epidemic Type Aftershock Sequences(ETAS)model.The stochastic components in the ETAS model are used as a proxy for possible earthquake triggered by external forces(fluids).The results show that the proportion of earthquakes triggered by external forces of Chang Island swarm increases gradually(from 31.9%to 63.5%)and then decreases.The latter stage of swarm development is mainly affected by the self-excitation of earthquakes,suggesting that the fluids play an important role in the development of the Chang Island swarm.However,the triggering intensity of fluids to microseismicity is divergent in different periods,which may be related to the process of fluid permeation.
基金supported by the National Natural Science Foundation of China(Grant Nos.62472149,62376089,62202147)Hubei Provincial Science and Technology Plan Project(2023BCB04100).
文摘Accurate traffic flow prediction has a profound impact on modern traffic management. Traffic flow has complex spatial-temporal correlations and periodicity, which poses difficulties for precise prediction. To address this problem, a Multi-head Self-attention and Spatial-Temporal Graph Convolutional Network (MSSTGCN) for multiscale traffic flow prediction is proposed. Firstly, to capture the hidden traffic periodicity of traffic flow, traffic flow is divided into three kinds of periods, including hourly, daily, and weekly data. Secondly, a graph attention residual layer is constructed to learn the global spatial features across regions. Local spatial-temporal dependence is captured by using a T-GCN module. Thirdly, a transformer layer is introduced to learn the long-term dependence in time. A position embedding mechanism is introduced to label position information for all traffic sequences. Thus, this multi-head self-attention mechanism can recognize the sequence order and allocate weights for different time nodes. Experimental results on four real-world datasets show that the MSSTGCN performs better than the baseline methods and can be successfully adapted to traffic prediction tasks.
基金supported by "Study on Strong Earthquake Risk in Southern Region of Longmenshan Fault,Huayingshan Fault and Border Area of Sichuan and Yunnan Provinces",Department of Earthquake Monitoring and Prediction,China Earthquake Administration
文摘In order to reveal the relationship between water injection in mine wells in the Zigong area and seismicity, we divide the historical earthquakes of ML ≥ 1.2 into 3 phases according to seismicity behavior, and the ETAS model parameters are then inversed by the POWELL method. The results show that phase 1 and 2, in which there is no water injection, have moderate-to-low ratio of background earthquakes （40 % - 50 % ）, and aftershocks are relatively less for a single earthquake sequence. In phase 3, where there is water injection, the aftershocks triggered by foreshocks dominate （ 93. 1% ）, and background earthquakes amount only to 6. 9 %, less than those of phase 1 and 2. The results conflict with the existing cognition. To resolve this problem, we propose that the occurrence ratio of background earthquakes in unit time, that is, the p value in ETAS model is used as an indicator of water injection triggered earthquakes. Compared to the first two phases, phase 3 has the largest u value, which illustrates that the water injection has an obvious triggering effect on earthquakes of this region.
文摘利用基于时-空传染型余震序列(Epidemic Type Aftershock Sequence,简称ETAS)模型的随机除丛法,重新审视了2008年5月12日汶川M_S8.0地震前可能存在的长期地震活动异常,研究了川滇地区背景地震活动特征,并评估了当前的强震危险状态.对川滇地区1 970年以来的M_L3.0以上的背景地震和丛集地震活动的研究结果表明,该地区地震丛集特征明显、时空分布很不均匀、地震序列常有前震事件.直接将概率值作为地震计数的权重,对地震丛集率空间分布图像分析表明,汶川M_S8.0地震前,龙门山断裂带中南段存在着长期、大范围的地震丛集率低值区,震前该段处于应力闭锁状态.对川滇地区地震从集率低值区内背景地震与全部地震的累积次数、b值和新定义的Δb等统计参量的分析表明,龙日坝与龙门山断裂带具有地震活动的关联性,川滇地区当前的强震潜在危险区可能是巧家地区和汶川M_S8.0地震破裂尚未穿越的龙门山断裂带南段.此外,还发现b值倾向于反映局部应力场变化,而△b能较为敏感地给出更大范围应力场的相对变化.
基金supported by the National Natural Science Foundation of China (71273105)the Fundamental Research Funds for the Central Universities,China (2013YB12)
文摘Macroscopic grasp of agricultural carbon emissions status, spatial-temporal characteristics as well as driving factors are the basic premise in further research on China’s agricultural carbon emissions. Based on 23 kinds of major carbon emission sources including agricultural materials inputs, paddy ifeld, soil and livestock breeding, this paper ifrstly calculated agricultural carbon emissions from 1995 to 2010, as well as 31 provinces and cities in 2010 in China. We then made a decomposed analysis to the driving factors of carbon emissions with logarithmic mean Divisia index （LMDI） model. The results show：（1） The amount of agricultural carbon emissions is 291.1691 million t in 2010. Compared with 249.5239 million t in 1995, it increased by 16.69%, in which, agricultural materials inputs, paddy ifeld, soil, enteric fermentation, and manure management accounted for 33.59, 22.03, 7.46, 17.53 and 19.39%of total agricultural carbon emissions, respectively. Although the amount exist ups and downs, it shows an overall trend of cyclical rise; （2） There is an obvious difference among regions：the amount of agricultural carbon emissions from top ten zones account for 56.68%, while 9.84%from last 10 zones. The traditional agricultural provinces, especially the major crop production areas are the main source regions. Based on the differences of carbon emission rations, 31 provinces and cities are divided into ifve types, namely agricultural materials dominant type, paddy ifeld dominant type, enteric fermentation dominant type, composite factors dominant type and balanced type. The agricultural carbon emissions intensity in west of China is the highest, followed by the central region, and the east zone is the lowest; （3） Compared with 1995, efifciency, labor and structure factors cut down carbon emissions by 65.78, 27.51 and 3.19%, respectively;while economy factor increase carbon emissions by 113.16%.
基金funded through the Special Fund for Agro-Scientific Research in the Public Interestthe Special Public Welfare Industry (agriculture) Research-Research and Demonstration of Fisheries Fishing Technology and Fishing Gear (No. 201203018)the National Natural Science Foundation of China (No. 31402350)
文摘Set-nets are common alongshore fishing gear used in Haizhou Bay, which rely on flow to catch fish. The catch per unit effort(CPUE) of set-net is affected by spatial-temporal and environmental factors but no research has been conducted on this subject. In this study, we used generalized additive models(GAMs) to explore the influence of spatial-temporal and environmental factors on CPUEs of species aggregated, small yellow croaker(Larimichthys polyactis), and octopus(Octopus variabilis) based on logbooks investigations conducted at 4 stations in an alongshore area of Haizhou Bay from 2011 to 2012. The results showed that all CPUEs exhibited significant spatial-temporal differences at various scales. Aggregated CPUE was high when the sea surface temperature(SST) was 15-18℃ and 20-23℃, which was mainly determined by life history traits of the octopus and small yellow croaker(optimal SSTs 14-17℃ and 19-24℃, respectively). Chlorophyll-a concentration had significant influences on the aggregated, small yellow croaker and octopus CPUEs at optimal ranges of 3.8-6.2 mg m^(-3), 4.2-4.8 mg m^(-3) and 4.5-5.5 mg m^(-3), respectively. Flow through the net had positive relationships with CPUEs. The approximate logarithmic trends in regression curves had a critical point of 2.5 Mm^3 d^(-1), which was the dividing point that differentiated whether the major factor affecting CPUEs was the flow velocity or the fishery resource. Our results from this study will help guide fishery production and improve catch rate of set-net fishing in Haizhou Bay.
基金Under the auspices of National Natural Science Foundation of China(No.41401182,41501173)Youth Fund for Humanities and Social Sciences of the Ministry of Education of China(No.19YJC630177)+2 种基金Natural Science Foundation of Heilongjiang Province(No.LH2019D008)University Nursing Program for Young Scholars with Creative Talents in Heilongjiang Province(No.UNPYSCT-2018194)Talent Introduction Project of Southwest University(No.SWU019020)。
文摘To comprehensively understand the law of urban-rural relationship and propose scientific measures of urban-rural coordinated development in Northeast China,this study uses the coupling coordination degree model and geographically and temporally weighted regression(GTWR)model to analyze the spatial-temporal patterns and the corresponding driving mechanisms of its urban-rural coordination since 1990.The results are as follows.First,the urban-rural coupling coordination degree in Northeast China was very low and improved slowly,but its stages of evolution is a good interpretation of the strategic arrangements of China's urbanization.Second,the urban-rural coupling coordination degree in Northeast China had spatial differences and was characterized by central polarization,converging on urban agglomeration,which was high in the south and low in the north.Moreover,the gap between the north and south weakened.Third,the spatial-temporal evolution of the urban-rural coordination relationship in Northeast China was influenced by pulling from the central cities,pushing from rural transformation,and government regulations.The influence intensity of the three mechanisms was weak,but the pulling from the central cities was stronger than that of the other two mechanisms.Furthermore,the spatial difference between the three mechanisms determines the spatial pattern and its evolution of the urban-rural coordination relationship in Northeast China.Fourth,to promote the development of urban-rural coordination in Northeast China,it is essential to advance urban-rural economic correlation,enhance the government^role in regulating and guiding,and adopt different policies for each region in Northeast China.
